There are good reasons to be skeptical that automation will unravel the labor market

When it comes to the threat of automation, I agree with Ryan Khurana: “From self-driving car crashes to failed workplace algorithms, many AI tools fail to perform simple tasks humans excel at, let alone far surpass us in every way.” Like myself, he is skeptical that automation will unravel the labor market, pointing out that “[The] conflation of what AI ‘may one day do’ with the much more mundane ‘what software can do today’ creates a powerful narrative around automation that accepts no refutation.”

Khurana marshals a number of examples to make this point:

Google needs to use human callers to impersonate its Duplex system on up to a quarter of calls, and Uber needs crowd-sourced labor to ensure its automated identification system remains fast, but admitting this makes them look less automated…

London-based investment firm MMC Ventures found that out of the 2,830 startups they identified as being “AI-focused” in Europe, 40 percent used no machine learning tools, whatsoever.

I’ve been collecting examples of the AI hype machine as well. Here are some of my favorites.

From Rodney Brooks comes this corrective: “Chris Urmson, the former leader of Google’s self-driving car project, once hoped that his son wouldn’t need a driver’s license because driverless cars would be so plentiful by 2020. Now the CEO of the self-driving startup Aurora, Urmson says that driverless cars will be slowly integrated onto our roads “over the next 30 to 50 years.”

Judea Pearl, a pioneer in statistics, said last year that “All the impressive achievements of deep learning amount to just curve fitting,” a technique that was developed decades ago.

Earlier this year, IBM shut down its Watson AI tool for drug discovery.

Mike Mallazzo said it this way: “The investors know it’s bullshit. When venture capitalists say they are looking to add ‘A.I. companies’ to their portfolio, what they really want is a technological moat built around access to uniquely valuable data. If it’s beneficial for companies to sprinkle in a little sex appeal and brand this as ‘A.I.,’ there’s no incentive to stop them from doing so.”

And there is the problem of cost:

• Google’s DeepMind lost roughly \$162 million in 2016.
• Facebook might have access to vast engineering resources and data about language, but still their chatbot project, M, fell short. According to one source familiar with the program, M never surpassed 30 percent automation.^[21]
• Ocado, the UK-based online supermarket, explained in late 2017 that the company would need to spend “an extra couple of million pounds” to hire software engineers to work on automating its warehouses.
• Facebook tried to largely automate their content moderation process, but had to pull back on the project and has instead upped the number of content moderators.
• After years of development, T-Mobile got rid of its robotic customer service lines.
• BMW and Daimler, long time competitors, have been pooling resources to create autonomous vehicles because of the astronomical costs involved.
• Uber recently raised a \$1 billion for its autonomous vehicles unit.
• McDonald’s purchased Silicon Valley VC-backed AI company Dynamic Yield, which was reportedly for over \$300 million, making it the fast-food giant’s largest acquisition since it bought Boston Market in 1999.
• Ford invested nearly a \$1 billion in Argo for its work in autonomous vehicles.
• Johnson & Johnson paid \$3.4 billion to pick up surgical robotics pioneer Auris Health, and its FDA-cleared Monarch platform in March.
• Total private investment into AI businesses in the United Kingdom exceeded £3.8 billion in 2018.
• SAS intends to spend \$1 billion on AI research in the coming three years.
• According to data from CB Insights, a record \$9.3 billion went to U.S.-based AI startups in 2018, an \$8.2 billion increase from the \$1.1 billion raised in 2013.

As I explained before, the large pecuniary costs in big data technologies don’t speak to the equally expensive task of overhauling management techniques to make the new systems work. New technologies can’t be seamlessly adopted within firms, they need management and process innovations to make the new data-driven methods profitable. And to be honest, we just aren’t there yet.